X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FSequence.java;h=a01d185c24267ea6368818907139a0ce9d5a6c86;hb=6ec47c00025dbeb48dba5f5db77754ec50b7a6b5;hp=e2fc057ed5947fefd2574623d9479554a55d33f3;hpb=582d39cb05dfbb5f956f74d4a97a17d9f63b0786;p=jalview.git diff --git a/src/jalview/datamodel/Sequence.java b/src/jalview/datamodel/Sequence.java index e2fc057..a01d185 100755 --- a/src/jalview/datamodel/Sequence.java +++ b/src/jalview/datamodel/Sequence.java @@ -1,106 +1,184 @@ /* -* Jalview - A Sequence Alignment Editor and Viewer -* Copyright (C) 2006 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle -* -* This program is free software; you can redistribute it and/or -* modify it under the terms of the GNU General Public License -* as published by the Free Software Foundation; either version 2 -* of the License, or (at your option) any later version. -* -* This program is distributed in the hope that it will be useful, -* but WITHOUT ANY WARRANTY; without even the implied warranty of -* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -* GNU General Public License for more details. -* -* You should have received a copy of the GNU General Public License -* along with this program; if not, write to the Free Software -* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA -*/ + * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$) + * Copyright (C) $$Year-Rel$$ The Jalview Authors + * + * This file is part of Jalview. + * + * Jalview is free software: you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation, either version 3 + * of the License, or (at your option) any later version. + * + * Jalview is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty + * of MERCHANTABILITY or FITNESS FOR A PARTICULAR + * PURPOSE. See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with Jalview. If not, see . + * The Jalview Authors are detailed in the 'AUTHORS' file. + */ package jalview.datamodel; -import java.awt.*; +import jalview.analysis.AlignSeq; +import jalview.api.DBRefEntryI; +import jalview.datamodel.features.SequenceFeatures; +import jalview.datamodel.features.SequenceFeaturesI; +import jalview.util.Comparison; +import jalview.util.DBRefUtils; +import jalview.util.MapList; +import jalview.util.StringUtils; + +import java.util.ArrayList; +import java.util.Arrays; +import java.util.BitSet; +import java.util.Collections; +import java.util.Enumeration; +import java.util.List; +import java.util.ListIterator; +import java.util.Vector; -import java.util.*; +import com.stevesoft.pat.Regex; +import fr.orsay.lri.varna.models.rna.RNA; /** - * DOCUMENT ME! - * + * + * Implements the SequenceI interface for a char[] based sequence object. + * * @author $author$ * @version $Revision$ */ -public class Sequence implements SequenceI +public class Sequence extends ASequence implements SequenceI { + private static final Regex limitrx = new Regex( + "[/][0-9]{1,}[-][0-9]{1,}$"); + + private static final Regex endrx = new Regex("[0-9]{1,}$"); + SequenceI datasetSequence; + String name; - private String sequence; + + private char[] sequence; + String description; + int start; + int end; - Vector pdbIds; + + Vector pdbIds; + String vamsasId; + DBRefEntry[] dbrefs; - /** This annotation is displayed below the alignment but the - * positions are tied to the residues of this sequence */ - Vector annotation; + RNA rna; + + /** + * This annotation is displayed below the alignment but the positions are tied + * to the residues of this sequence + * + * TODO: change to List<> + */ + Vector annotation; + + /** + * The index of the sequence in a MSA + */ + int index = -1; + + private SequenceFeatures sequenceFeatureStore; - /** DOCUMENT ME!! */ - public SequenceFeature[] sequenceFeatures; + /* + * A cursor holding the approximate current view position to the sequence, + * as determined by findIndex or findPosition or findPositions. + * Using a cursor as a hint allows these methods to be more performant for + * large sequences. + */ + private SequenceCursor cursor; - /** This array holds hidden sequences - * of which this sequence is the representitive member of a group + /* + * A number that should be incremented whenever the sequence is edited. + * If the value matches the cursor token, then we can trust the cursor, + * if not then it should be recomputed. */ - SequenceGroup hiddenSequences; + private int changeCount; /** * Creates a new Sequence object. - * - * @param name DOCUMENT ME! - * @param sequence DOCUMENT ME! - * @param start DOCUMENT ME! - * @param end DOCUMENT ME! + * + * @param name + * display name string + * @param sequence + * string to form a possibly gapped sequence out of + * @param start + * first position of non-gap residue in the sequence + * @param end + * last position of ungapped residues (nearly always only used for + * display purposes) */ public Sequence(String name, String sequence, int start, int end) { - this.name = name; - this.sequence = sequence; - this.start = start; - this.end = end; + this(); + initSeqAndName(name, sequence.toCharArray(), start, end); + } - parseId(); + public Sequence(String name, char[] sequence, int start, int end) + { + this(); + initSeqAndName(name, sequence, start, end); + } + /** + * Stage 1 constructor - assign name, sequence, and set start and end fields. + * start and end are updated values from name2 if it ends with /start-end + * + * @param name2 + * @param sequence2 + * @param start2 + * @param end2 + */ + protected void initSeqAndName(String name2, char[] sequence2, int start2, + int end2) + { + this.name = name2; + this.sequence = sequence2; + this.start = start2; + this.end = end2; + parseId(); checkValidRange(); } - com.stevesoft.pat.Regex limitrx = new com.stevesoft.pat.Regex( - "[/][0-9]{1,}[-][0-9]{1,}$"); - com.stevesoft.pat.Regex endrx = new com.stevesoft.pat.Regex( - "[0-9]{1,}$"); - void parseId() { - // Does sequence have the /start-end signiature? + if (name == null) + { + System.err + .println("POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor."); + name = ""; + } + // Does sequence have the /start-end signature? if (limitrx.search(name)) { name = limitrx.left(); endrx.search(limitrx.stringMatched()); setStart(Integer.parseInt(limitrx.stringMatched().substring(1, - endrx.matchedFrom() - 1))); + endrx.matchedFrom() - 1))); setEnd(Integer.parseInt(endrx.stringMatched())); } } void checkValidRange() { - if (end < 1) + // Note: JAL-774 : + // http://issues.jalview.org/browse/JAL-774?focusedCommentId=11239&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-11239 { int endRes = 0; - char ch; - for (int j = 0; j < sequence.length(); j++) + for (int j = 0; j < sequence.length; j++) { - ch = sequence.charAt(j); - if (!jalview.util.Comparison.isGap( (ch))) + if (!jalview.util.Comparison.isGap(sequence[j])) { endRes++; } @@ -110,16 +188,29 @@ public class Sequence implements SequenceI endRes += start - 1; } - this.end = endRes; + if (end < endRes) + { + end = endRes; + } } } /** + * default constructor + */ + private Sequence() + { + sequenceFeatureStore = new SequenceFeatures(); + } + + /** * Creates a new Sequence object. - * - * @param name DOCUMENT ME! - * @param sequence DOCUMENT ME! + * + * @param name + * DOCUMENT ME! + * @param sequence + * DOCUMENT ME! */ public Sequence(String name, String sequence) { @@ -127,128 +218,229 @@ public class Sequence implements SequenceI } /** - * Creates a new Sequence object. - * - * @param seq DOCUMENT ME! + * Creates a new Sequence object with new AlignmentAnnotations but inherits + * any existing dataset sequence reference. If non exists, everything is + * copied. + * + * @param seq + * if seq is a dataset sequence, behaves like a plain old copy + * constructor */ public Sequence(SequenceI seq) { - this(seq.getName(), seq.getSequence(), seq.getStart(), seq.getEnd()); + this(seq, seq.getAnnotation()); } /** - * DOCUMENT ME! - * - * @param v DOCUMENT ME! + * Create a new sequence object with new features, DBRefEntries, and PDBIds + * but inherits any existing dataset sequence reference, and duplicate of any + * annotation that is present in the given annotation array. + * + * @param seq + * the sequence to be copied + * @param alAnnotation + * an array of annotation including some associated with seq */ - public void setSequenceFeatures(SequenceFeature[] features) + public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation) { - sequenceFeatures = features; + this(); + initSeqFrom(seq, alAnnotation); } - public synchronized void addSequenceFeature(SequenceFeature sf) + /** + * does the heavy lifting when cloning a dataset sequence, or coping data from + * dataset to a new derived sequence. + * + * @param seq + * - source of attributes. + * @param alAnnotation + * - alignment annotation present on seq that should be copied onto + * this sequence + */ + protected void initSeqFrom(SequenceI seq, + AlignmentAnnotation[] alAnnotation) { - if (sequenceFeatures == null) + char[] oseq = seq.getSequence(); // returns a copy of the array + initSeqAndName(seq.getName(), oseq, seq.getStart(), seq.getEnd()); + + description = seq.getDescription(); + if (seq != datasetSequence) { - sequenceFeatures = new SequenceFeature[0]; + setDatasetSequence(seq.getDatasetSequence()); } - - for (int i = 0; i < sequenceFeatures.length; i++) + + /* + * only copy DBRefs and seqfeatures if we really are a dataset sequence + */ + if (datasetSequence == null) { - if (sequenceFeatures[i].equals(sf)) + if (seq.getDBRefs() != null) { - return; + DBRefEntry[] dbr = seq.getDBRefs(); + for (int i = 0; i < dbr.length; i++) + { + addDBRef(new DBRefEntry(dbr[i])); + } } - } - SequenceFeature[] temp = new SequenceFeature[sequenceFeatures.length + 1]; - System.arraycopy(sequenceFeatures, 0, temp, 0, sequenceFeatures.length); - temp[sequenceFeatures.length] = sf; + /* + * make copies of any sequence features + */ + for (SequenceFeature sf : seq.getSequenceFeatures()) + { + addSequenceFeature(new SequenceFeature(sf)); + } + } - sequenceFeatures = temp; + if (seq.getAnnotation() != null) + { + AlignmentAnnotation[] sqann = seq.getAnnotation(); + for (int i = 0; i < sqann.length; i++) + { + if (sqann[i] == null) + { + continue; + } + boolean found = (alAnnotation == null); + if (!found) + { + for (int apos = 0; !found && apos < alAnnotation.length; apos++) + { + found = (alAnnotation[apos] == sqann[i]); + } + } + if (found) + { + // only copy the given annotation + AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]); + addAlignmentAnnotation(newann); + } + } + } + if (seq.getAllPDBEntries() != null) + { + Vector ids = seq.getAllPDBEntries(); + for (PDBEntry pdb : ids) + { + this.addPDBId(new PDBEntry(pdb)); + } + } } - public void deleteFeature(SequenceFeature sf) + @Override + public void setSequenceFeatures(List features) { - if(sequenceFeatures==null) + if (datasetSequence != null) + { + datasetSequence.setSequenceFeatures(features); return; + } + sequenceFeatureStore = new SequenceFeatures(features); + } - int index=0; - for (index = 0; index < sequenceFeatures.length; index++) + @Override + public synchronized boolean addSequenceFeature(SequenceFeature sf) + { + if (sf.getType() == null) { - if (sequenceFeatures[index].equals(sf)) - { - break; - } + System.err.println("SequenceFeature type may not be null: " + + sf.toString()); + return false; } + if (datasetSequence != null) + { + return datasetSequence.addSequenceFeature(sf); + } - if(index==sequenceFeatures.length) - return; + return sequenceFeatureStore.add(sf); + } - int sfLength = sequenceFeatures.length; - if(sfLength<2) + @Override + public void deleteFeature(SequenceFeature sf) + { + if (datasetSequence != null) { - sequenceFeatures = null; + datasetSequence.deleteFeature(sf); } else { - SequenceFeature[] temp = new SequenceFeature[sfLength-1]; - System.arraycopy(sequenceFeatures, 0, temp, 0, index); - - if(index getSequenceFeatures() + { + if (datasetSequence != null) + { + return datasetSequence.getSequenceFeatures(); + } + return sequenceFeatureStore.getAllFeatures(); + } + + @Override + public SequenceFeaturesI getFeatures() { - return sequenceFeatures; + return datasetSequence != null ? datasetSequence.getFeatures() + : sequenceFeatureStore; } - public void addPDBId(PDBEntry entry) + @Override + public boolean addPDBId(PDBEntry entry) { if (pdbIds == null) - pdbIds = new Vector(); + { + pdbIds = new Vector(); + pdbIds.add(entry); + return true; + } + for (PDBEntry pdbe : pdbIds) + { + if (pdbe.updateFrom(entry)) + { + return false; + } + } pdbIds.addElement(entry); + return true; } /** * DOCUMENT ME! - * - * @param id DOCUMENT ME! + * + * @param id + * DOCUMENT ME! */ - public void setPDBId(Vector id) + @Override + public void setPDBId(Vector id) { pdbIds = id; } /** * DOCUMENT ME! - * + * * @return DOCUMENT ME! */ - public Vector getPDBId() + @Override + public Vector getAllPDBEntries() { - return pdbIds; + return pdbIds == null ? new Vector() : pdbIds; } /** * DOCUMENT ME! - * + * * @return DOCUMENT ME! */ + @Override public String getDisplayId(boolean jvsuffix) { StringBuffer result = new StringBuffer(name); @@ -262,9 +454,11 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * - * @param name DOCUMENT ME! + * + * @param name + * DOCUMENT ME! */ + @Override public void setName(String name) { this.name = name; @@ -273,9 +467,10 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * + * * @return DOCUMENT ME! */ + @Override public String getName() { return this.name; @@ -283,9 +478,11 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * - * @param start DOCUMENT ME! + * + * @param start + * DOCUMENT ME! */ + @Override public void setStart(int start) { this.start = start; @@ -293,9 +490,10 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * + * * @return DOCUMENT ME! */ + @Override public int getStart() { return this.start; @@ -303,9 +501,11 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * - * @param end DOCUMENT ME! + * + * @param end + * DOCUMENT ME! */ + @Override public void setEnd(int end) { this.end = end; @@ -313,9 +513,10 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * + * * @return DOCUMENT ME! */ + @Override public int getEnd() { return this.end; @@ -323,93 +524,119 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * + * * @return DOCUMENT ME! */ + @Override public int getLength() { - return this.sequence.length(); + return this.sequence.length; } /** * DOCUMENT ME! - * - * @param seq DOCUMENT ME! + * + * @param seq + * DOCUMENT ME! */ + @Override public void setSequence(String seq) { - this.sequence = seq; + this.sequence = seq.toCharArray(); checkValidRange(); + sequenceChanged(); } - /** - * DOCUMENT ME! - * - * @return DOCUMENT ME! - */ - public String getSequence() + @Override + public String getSequenceAsString() { - return this.sequence; + return new String(sequence); } - /** - * DOCUMENT ME! - * - * @param start DOCUMENT ME! - * @param end DOCUMENT ME! - * - * @return DOCUMENT ME! + @Override + public String getSequenceAsString(int start, int end) + { + return new String(getSequence(start, end)); + } + + @Override + public char[] getSequence() + { + // return sequence; + return sequence == null ? null : Arrays.copyOf(sequence, + sequence.length); + } + + /* + * (non-Javadoc) + * + * @see jalview.datamodel.SequenceI#getSequence(int, int) */ - public String getSequence(int start, int end) + @Override + public char[] getSequence(int start, int end) { - // JBPNote - left to user to pad the result here (TODO:Decide on this policy) - if (start >= sequence.length()) + if (start < 0) + { + start = 0; + } + // JBPNote - left to user to pad the result here (TODO:Decide on this + // policy) + if (start >= sequence.length) { - return ""; + return new char[0]; } - if (end >= sequence.length()) + if (end >= sequence.length) { - end = sequence.length(); + end = sequence.length; } - return this.sequence.substring(start, end); + char[] reply = new char[end - start]; + System.arraycopy(sequence, start, reply, 0, end - start); + + return reply; } - /** - * make a new Sequence object from start to end (including gaps) over this seqeunce - * @param start int - * @param end int - * @return SequenceI - */ + @Override public SequenceI getSubSequence(int start, int end) { if (start < 0) + { start = 0; - String seq = getSequence(start, end); - if (seq == "") + } + char[] seq = getSequence(start, end); + if (seq.length == 0) + { return null; + } int nstart = findPosition(start); int nend = findPosition(end) - 1; // JBPNote - this is an incomplete copy. SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend); nseq.setDescription(description); - nseq.setDatasetSequence(getDatasetSequence()); + if (datasetSequence != null) + { + nseq.setDatasetSequence(datasetSequence); + } + else + { + nseq.setDatasetSequence(this); + } return nseq; } /** - * DOCUMENT ME! - * - * @param i DOCUMENT ME! - * - * @return DOCUMENT ME! + * Returns the character of the aligned sequence at the given position (base + * zero), or space if the position is not within the sequence's bounds + * + * @return */ + @Override public char getCharAt(int i) { - if (i < sequence.length()) + if (i >= 0 && i < sequence.length) { - return sequence.charAt(i); + return sequence[i]; } else { @@ -419,9 +646,11 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * - * @param desc DOCUMENT ME! + * + * @param desc + * DOCUMENT ME! */ + @Override public void setDescription(String desc) { this.description = desc; @@ -429,258 +658,1245 @@ public class Sequence implements SequenceI /** * DOCUMENT ME! - * + * * @return DOCUMENT ME! */ + @Override public String getDescription() { return this.description; } /** - * DOCUMENT ME! - * - * @param pos DOCUMENT ME! - * - * @return DOCUMENT ME! + * {@inheritDoc} */ + @Override public int findIndex(int pos) { - // returns the alignment position for a residue + /* + * use a valid, hopefully nearby, cursor if available + */ + if (isValidCursor(cursor)) + { + return findIndex(pos, cursor); + } + int j = start; int i = 0; + int startColumn = 0; - while ( (i < sequence.length()) && (j <= end) && (j <= pos)) + /* + * traverse sequence from the start counting gaps; make a note of + * the column of the first residue to save in the cursor + */ + while ((i < sequence.length) && (j <= end) && (j <= pos)) { - if (!jalview.util.Comparison.isGap(sequence.charAt(i))) + if (!Comparison.isGap(sequence[i])) { + if (j == start) + { + startColumn = i; + } j++; } - i++; } - if ( (j == end) && (j < pos)) + if (j == end && j < pos) { return end + 1; } - else + + updateCursor(pos, i, startColumn); + return i; + } + + /** + * Updates the cursor to the latest found residue and column position + * + * @param residuePos + * (start..) + * @param column + * (1..) + * @param startColumn + * column position of the first sequence residue + */ + protected void updateCursor(int residuePos, int column, int startColumn) + { + /* + * preserve end residue column provided cursor was valid + */ + int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0; + if (residuePos == this.end) { - return i; + endColumn = column; } + + cursor = new SequenceCursor(this, residuePos, column, startColumn, + endColumn, this.changeCount); } /** - * Returns the sequence position for an alignment position - * - * @param i column index in alignment (from 1) - * - * @return residue number for residue (left of and) nearest ith column + * Answers the aligned column position (1..) for the given residue position + * (start..) given a 'hint' of a residue/column location in the neighbourhood. + * The hint may be left of, at, or to the right of the required position. + * + * @param pos + * @param curs + * @return */ - public int findPosition(int i) + protected int findIndex(int pos, SequenceCursor curs) { - int j = 0; - int pos = start; - int seqlen = sequence.length(); - while ( (j < i) && (j < seqlen)) + if (!isValidCursor(curs)) + { + /* + * wrong or invalidated cursor, compute de novo + */ + return findIndex(pos); + } + + if (curs.residuePosition == pos) + { + return curs.columnPosition; + } + + /* + * move left or right to find pos from hint.position + */ + int col = curs.columnPosition - 1; // convert from base 1 to 0-based array + // index + int newPos = curs.residuePosition; + int delta = newPos > pos ? -1 : 1; + + while (newPos != pos) { - if (!jalview.util.Comparison.isGap( (sequence.charAt(j)))) + col += delta; // shift one column left or right + if (col < 0 || col == sequence.length) { - pos++; + break; + } + if (!Comparison.isGap(sequence[col])) + { + newPos += delta; } - - j++; } - return pos; + col++; // convert back to base 1 + updateCursor(pos, col, curs.firstColumnPosition); + + return col; } /** - * Returns an int array where indices correspond to each residue in the sequence and the element value gives its position in the alignment - * - * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no residues in SequenceI object + * {@inheritDoc} */ - public int[] gapMap() + @Override + public int findPosition(final int column) { - String seq = jalview.analysis.AlignSeq.extractGaps(jalview.util.Comparison. - GapChars, sequence); - int[] map = new int[seq.length()]; + /* + * use a valid, hopefully nearby, cursor if available + */ + if (isValidCursor(cursor)) + { + return findPosition(column + 1, cursor); + } + + // TODO recode this more naturally i.e. count residues only + // as they are found, not 'in anticipation' + + /* + * traverse the sequence counting gaps; note the column position + * of the first residue, to save in the cursor + */ + int firstResidueColumn = 0; + int lastPosFound = 0; + int lastPosFoundColumn = 0; + int seqlen = sequence.length; + + if (seqlen > 0 && !Comparison.isGap(sequence[0])) + { + lastPosFound = start; + lastPosFoundColumn = 0; + } + int j = 0; - int p = 0; + int pos = start; - while (j < sequence.length()) + while (j < column && j < seqlen) { - if (!jalview.util.Comparison.isGap(sequence.charAt(j))) + if (!Comparison.isGap(sequence[j])) { - map[p++] = j; + lastPosFound = pos; + lastPosFoundColumn = j; + if (pos == this.start) + { + firstResidueColumn = j; + } + pos++; } - j++; } + if (j < seqlen && !Comparison.isGap(sequence[j])) + { + lastPosFound = pos; + lastPosFoundColumn = j; + if (pos == this.start) + { + firstResidueColumn = j; + } + } - return map; + /* + * update the cursor to the last residue position found (if any) + * (converting column position to base 1) + */ + if (lastPosFound != 0) + { + updateCursor(lastPosFound, lastPosFoundColumn + 1, + firstResidueColumn + 1); + } + + return pos; } /** - * DOCUMENT ME! - * - * @param i DOCUMENT ME! + * Answers true if the given cursor is not null, is for this sequence object, + * and has a token value that matches this object's changeCount, else false. + * This allows us to ignore a cursor as 'stale' if the sequence has been + * modified since the cursor was created. + * + * @param curs + * @return */ - public void deleteCharAt(int i) + protected boolean isValidCursor(SequenceCursor curs) { - if (i >= sequence.length()) + if (curs == null || curs.sequence != this || curs.token != changeCount) { - return; + return false; } - - sequence = sequence.substring(0, i) + sequence.substring(i + 1); + /* + * sanity check against range + */ + if (curs.columnPosition < 0 || curs.columnPosition > sequence.length) + { + return false; + } + if (curs.residuePosition < start || curs.residuePosition > end) + { + return false; + } + return true; } /** - * DOCUMENT ME! - * - * @param i DOCUMENT ME! - * @param j DOCUMENT ME! + * Answers the sequence position (start..) for the given aligned column + * position (1..), given a hint of a cursor in the neighbourhood. The cursor + * may lie left of, at, or to the right of the column position. + * + * @param col + * @param curs + * @return */ - public void deleteChars(int i, int j) + protected int findPosition(final int col, SequenceCursor curs) { - if (i >= sequence.length()) + if (!isValidCursor(curs)) { - return; + /* + * wrong or invalidated cursor, compute de novo + */ + return findPosition(col - 1);// ugh back to base 0 } - if (j >= sequence.length()) + if (curs.columnPosition == col) { - sequence = sequence.substring(0, i); + cursor = curs; // in case this method becomes public + return curs.residuePosition; // easy case :-) } - else + + if (curs.lastColumnPosition > 0 && curs.lastColumnPosition < col) { - sequence = sequence.substring(0, i) + sequence.substring(j); + /* + * sequence lies entirely to the left of col + * - return last residue + 1 + */ + return end + 1; } - } - /** - * DOCUMENT ME! - * - * @param i DOCUMENT ME! - * @param c DOCUMENT ME! - * @param chop DOCUMENT ME! - */ - public void insertCharAt(int i, int length, char c) - { - StringBuffer tmp; + if (curs.firstColumnPosition > 0 && curs.firstColumnPosition > col) + { + /* + * sequence lies entirely to the right of col + * - return first residue + */ + return start; + } - if (i >= sequence.length()) + // todo could choose closest to col out of column, + // firstColumnPosition, lastColumnPosition as a start point + + /* + * move left or right to find pos from cursor position + */ + int firstResidueColumn = curs.firstColumnPosition; + int column = curs.columnPosition - 1; // to base 0 + int newPos = curs.residuePosition; + int delta = curs.columnPosition > col ? -1 : 1; + boolean gapped = false; + int lastFoundPosition = curs.residuePosition; + int lastFoundPositionColumn = curs.columnPosition; + + while (column != col - 1) { - tmp = new StringBuffer(sequence); + column += delta; // shift one column left or right + if (column < 0 || column == sequence.length) + { + break; + } + gapped = Comparison.isGap(sequence[column]); + if (!gapped) + { + newPos += delta; + lastFoundPosition = newPos; + lastFoundPositionColumn = column + 1; + if (lastFoundPosition == this.start) + { + firstResidueColumn = column + 1; + } + } } - else - tmp = new StringBuffer(sequence.substring(0, i)); - while (length > 0) + if (cursor == null || lastFoundPosition != cursor.residuePosition) { - tmp.append(c); - length--; + updateCursor(lastFoundPosition, lastFoundPositionColumn, + firstResidueColumn); } - if (i < sequence.length()) + /* + * hack to give position to the right if on a gap + * or beyond the length of the sequence (see JAL-2562) + */ + if (delta > 0 && (gapped || column >= sequence.length)) { - tmp.append(sequence.substring(i)); + newPos++; } - sequence = tmp.toString(); + return newPos; } - public void insertCharAt(int i, char c) + /** + * {@inheritDoc} + */ + @Override + public Range findPositions(int fromColumn, int toColumn) { - insertCharAt(i, 1, c); - } + if (toColumn < fromColumn || fromColumn < 1) + { + return null; + } - public String getVamsasId() + /* + * find the first non-gapped position, if any + */ + int firstPosition = 0; + int col = fromColumn - 1; + int length = sequence.length; + while (col < length && col < toColumn) + { + if (!Comparison.isGap(sequence[col])) + { + firstPosition = findPosition(col++); + break; + } + col++; + } + + if (firstPosition == 0) + { + return null; + } + + /* + * find the last non-gapped position + */ + int lastPosition = firstPosition; + while (col < length && col < toColumn) + { + if (!Comparison.isGap(sequence[col++])) + { + lastPosition++; + } + } + + return new Range(firstPosition, lastPosition); + } + + /** + * Returns an int array where indices correspond to each residue in the + * sequence and the element value gives its position in the alignment + * + * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no + * residues in SequenceI object + */ + @Override + public int[] gapMap() + { + String seq = jalview.analysis.AlignSeq.extractGaps( + jalview.util.Comparison.GapChars, new String(sequence)); + int[] map = new int[seq.length()]; + int j = 0; + int p = 0; + + while (j < sequence.length) + { + if (!jalview.util.Comparison.isGap(sequence[j])) + { + map[p++] = j; + } + + j++; + } + + return map; + } + + @Override + public int[] findPositionMap() + { + int map[] = new int[sequence.length]; + int j = 0; + int pos = start; + int seqlen = sequence.length; + while ((j < seqlen)) + { + map[j] = pos; + if (!jalview.util.Comparison.isGap(sequence[j])) + { + pos++; + } + + j++; + } + return map; + } + + @Override + public List getInsertions() + { + ArrayList map = new ArrayList(); + int lastj = -1, j = 0; + int pos = start; + int seqlen = sequence.length; + while ((j < seqlen)) + { + if (jalview.util.Comparison.isGap(sequence[j])) + { + if (lastj == -1) + { + lastj = j; + } + } + else + { + if (lastj != -1) + { + map.add(new int[] { lastj, j - 1 }); + lastj = -1; + } + } + j++; + } + if (lastj != -1) + { + map.add(new int[] { lastj, j - 1 }); + lastj = -1; + } + return map; + } + + @Override + public BitSet getInsertionsAsBits() + { + BitSet map = new BitSet(); + int lastj = -1, j = 0; + int pos = start; + int seqlen = sequence.length; + while ((j < seqlen)) + { + if (jalview.util.Comparison.isGap(sequence[j])) + { + if (lastj == -1) + { + lastj = j; + } + } + else + { + if (lastj != -1) + { + map.set(lastj, j); + lastj = -1; + } + } + j++; + } + if (lastj != -1) + { + map.set(lastj, j); + lastj = -1; + } + return map; + } + + @Override + public void deleteChars(int i, int j) + { + int newstart = start, newend = end; + if (i >= sequence.length || i < 0) + { + return; + } + + char[] tmp = StringUtils.deleteChars(sequence, i, j); + boolean createNewDs = false; + // TODO: take a (second look) at the dataset creation validation method for + // the very large sequence case + int eindex = -1, sindex = -1; + boolean ecalc = false, scalc = false; + for (int s = i; s < j; s++) + { + if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23) + { + if (createNewDs) + { + newend--; + } + else + { + if (!scalc) + { + sindex = findIndex(start) - 1; + scalc = true; + } + if (sindex == s) + { + // delete characters including start of sequence + newstart = findPosition(j); + break; // don't need to search for any more residue characters. + } + else + { + // delete characters after start. + if (!ecalc) + { + eindex = findIndex(end) - 1; + ecalc = true; + } + if (eindex < j) + { + // delete characters at end of sequence + newend = findPosition(i - 1); + break; // don't need to search for any more residue characters. + } + else + { + createNewDs = true; + newend--; // decrease end position by one for the deleted residue + // and search further + } + } + } + } + } + // deletion occured in the middle of the sequence + if (createNewDs && this.datasetSequence != null) + { + // construct a new sequence + Sequence ds = new Sequence(datasetSequence); + // TODO: remove any non-inheritable properties ? + // TODO: create a sequence mapping (since there is a relation here ?) + ds.deleteChars(i, j); + datasetSequence = ds; + } + start = newstart; + end = newend; + sequence = tmp; + sequenceChanged(); + } + + @Override + public void insertCharAt(int i, int length, char c) + { + char[] tmp = new char[sequence.length + length]; + + if (i >= sequence.length) + { + System.arraycopy(sequence, 0, tmp, 0, sequence.length); + i = sequence.length; + } + else + { + System.arraycopy(sequence, 0, tmp, 0, i); + } + + int index = i; + while (length > 0) + { + tmp[index++] = c; + length--; + } + + if (i < sequence.length) + { + System.arraycopy(sequence, i, tmp, index, sequence.length - i); + } + + sequence = tmp; + sequenceChanged(); + } + + @Override + public void insertCharAt(int i, char c) + { + insertCharAt(i, 1, c); + } + + @Override + public String getVamsasId() { return vamsasId; } + @Override public void setVamsasId(String id) { vamsasId = id; } - public void setDBRef(DBRefEntry[] dbref) + @Override + public void setDBRefs(DBRefEntry[] dbref) { + if (dbrefs == null && datasetSequence != null + && this != datasetSequence) + { + datasetSequence.setDBRefs(dbref); + return; + } dbrefs = dbref; + if (dbrefs != null) + { + DBRefUtils.ensurePrimaries(this); + } } - public DBRefEntry[] getDBRef() + @Override + public DBRefEntry[] getDBRefs() { + if (dbrefs == null && datasetSequence != null + && this != datasetSequence) + { + return datasetSequence.getDBRefs(); + } return dbrefs; } + @Override public void addDBRef(DBRefEntry entry) { + if (datasetSequence != null) + { + datasetSequence.addDBRef(entry); + return; + } + if (dbrefs == null) + { dbrefs = new DBRefEntry[0]; + } - DBRefEntry[] temp = new DBRefEntry[dbrefs.length + 1]; - System.arraycopy(dbrefs, 0, temp, 0, dbrefs.length); + for (DBRefEntryI dbr : dbrefs) + { + if (dbr.updateFrom(entry)) + { + /* + * found a dbref that either matched, or could be + * updated from, the new entry - no need to add it + */ + return; + } + } + /* + * extend the array to make room for one more + */ + // TODO use an ArrayList instead + int j = dbrefs.length; + DBRefEntry[] temp = new DBRefEntry[j + 1]; + System.arraycopy(dbrefs, 0, temp, 0, j); temp[temp.length - 1] = entry; dbrefs = temp; + + DBRefUtils.ensurePrimaries(this); } + @Override public void setDatasetSequence(SequenceI seq) { + if (seq == this) + { + throw new IllegalArgumentException( + "Implementation Error: self reference passed to SequenceI.setDatasetSequence"); + } + if (seq != null && seq.getDatasetSequence() != null) + { + throw new IllegalArgumentException( + "Implementation error: cascading dataset sequences are not allowed."); + } datasetSequence = seq; } + @Override public SequenceI getDatasetSequence() { return datasetSequence; } + @Override public AlignmentAnnotation[] getAnnotation() { - if (annotation == null) - return null; - - AlignmentAnnotation[] ret = new AlignmentAnnotation[annotation.size()]; - for (int r = 0; r < ret.length; r++) - ret[r] = (AlignmentAnnotation) annotation.elementAt(r); + return annotation == null ? null : annotation + .toArray(new AlignmentAnnotation[annotation.size()]); + } - return ret; + @Override + public boolean hasAnnotation(AlignmentAnnotation ann) + { + return annotation == null ? false : annotation.contains(ann); } + @Override public void addAlignmentAnnotation(AlignmentAnnotation annotation) { if (this.annotation == null) - this.annotation = new Vector(); + { + this.annotation = new Vector(); + } + if (!this.annotation.contains(annotation)) + { + this.annotation.addElement(annotation); + } + annotation.setSequenceRef(this); + } - this.annotation.addElement(annotation); + @Override + public void removeAlignmentAnnotation(AlignmentAnnotation annotation) + { + if (this.annotation != null) + { + this.annotation.removeElement(annotation); + if (this.annotation.size() == 0) + { + this.annotation = null; + } + } } - public SequenceGroup getHiddenSequences() + /** + * test if this is a valid candidate for another sequence's dataset sequence. + * + */ + private boolean isValidDatasetSequence() { - return hiddenSequences; + if (datasetSequence != null) + { + return false; + } + for (int i = 0; i < sequence.length; i++) + { + if (jalview.util.Comparison.isGap(sequence[i])) + { + return false; + } + } + return true; } - public void addHiddenSequence(SequenceI seq) + @Override + public SequenceI deriveSequence() { - if (hiddenSequences == null) + Sequence seq = null; + if (datasetSequence == null) { - hiddenSequences = new SequenceGroup(); + if (isValidDatasetSequence()) + { + // Use this as dataset sequence + seq = new Sequence(getName(), "", 1, -1); + seq.setDatasetSequence(this); + seq.initSeqFrom(this, getAnnotation()); + return seq; + } + else + { + // Create a new, valid dataset sequence + createDatasetSequence(); + } } - hiddenSequences.addSequence(seq, false); + return new Sequence(this); } - public void showHiddenSequence(SequenceI seq) + private boolean _isNa; + + private int _seqhash = 0; + + /** + * Answers false if the sequence is more than 85% nucleotide (ACGTU), else + * true + */ + @Override + public boolean isProtein() { - hiddenSequences.deleteSequence(seq, false); - if (hiddenSequences.getSize(false) < 1) + if (datasetSequence != null) { - hiddenSequences = null; + return datasetSequence.isProtein(); } + if (_seqhash != sequence.hashCode()) + { + _seqhash = sequence.hashCode(); + _isNa = Comparison.isNucleotide(this); + } + return !_isNa; + }; + + /* + * (non-Javadoc) + * + * @see jalview.datamodel.SequenceI#createDatasetSequence() + */ + @Override + public SequenceI createDatasetSequence() + { + if (datasetSequence == null) + { + Sequence dsseq = new Sequence(getName(), AlignSeq.extractGaps( + jalview.util.Comparison.GapChars, getSequenceAsString()), + getStart(), getEnd()); + + datasetSequence = dsseq; + + dsseq.setDescription(description); + // move features and database references onto dataset sequence + dsseq.sequenceFeatureStore = sequenceFeatureStore; + sequenceFeatureStore = null; + dsseq.dbrefs = dbrefs; + dbrefs = null; + // TODO: search and replace any references to this sequence with + // references to the dataset sequence in Mappings on dbref + dsseq.pdbIds = pdbIds; + pdbIds = null; + datasetSequence.updatePDBIds(); + if (annotation != null) + { + // annotation is cloned rather than moved, to preserve what's currently + // on the alignment + for (AlignmentAnnotation aa : annotation) + { + AlignmentAnnotation _aa = new AlignmentAnnotation(aa); + _aa.sequenceRef = datasetSequence; + _aa.adjustForAlignment(); // uses annotation's own record of + // sequence-column mapping + datasetSequence.addAlignmentAnnotation(_aa); + } + } + } + return datasetSequence; } -} + /* + * (non-Javadoc) + * + * @see + * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[] + * annotations) + */ + @Override + public void setAlignmentAnnotation(AlignmentAnnotation[] annotations) + { + if (annotation != null) + { + annotation.removeAllElements(); + } + if (annotations != null) + { + for (int i = 0; i < annotations.length; i++) + { + if (annotations[i] != null) + { + addAlignmentAnnotation(annotations[i]); + } + } + } + } + + @Override + public AlignmentAnnotation[] getAnnotation(String label) + { + if (annotation == null || annotation.size() == 0) + { + return null; + } + + Vector subset = new Vector(); + Enumeration e = annotation.elements(); + while (e.hasMoreElements()) + { + AlignmentAnnotation ann = e.nextElement(); + if (ann.label != null && ann.label.equals(label)) + { + subset.addElement(ann); + } + } + if (subset.size() == 0) + { + return null; + } + AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()]; + int i = 0; + e = subset.elements(); + while (e.hasMoreElements()) + { + anns[i++] = e.nextElement(); + } + subset.removeAllElements(); + return anns; + } + @Override + public boolean updatePDBIds() + { + if (datasetSequence != null) + { + // TODO: could merge DBRefs + return datasetSequence.updatePDBIds(); + } + if (dbrefs == null || dbrefs.length == 0) + { + return false; + } + boolean added = false; + for (DBRefEntry dbr : dbrefs) + { + if (DBRefSource.PDB.equals(dbr.getSource())) + { + /* + * 'Add' any PDB dbrefs as a PDBEntry - add is only performed if the + * PDB id is not already present in a 'matching' PDBEntry + * Constructor parses out a chain code if appended to the accession id + * (a fudge used to 'store' the chain code in the DBRef) + */ + PDBEntry pdbe = new PDBEntry(dbr); + added |= addPDBId(pdbe); + } + } + return added; + } + + @Override + public void transferAnnotation(SequenceI entry, Mapping mp) + { + if (datasetSequence != null) + { + datasetSequence.transferAnnotation(entry, mp); + return; + } + if (entry.getDatasetSequence() != null) + { + transferAnnotation(entry.getDatasetSequence(), mp); + return; + } + // transfer any new features from entry onto sequence + if (entry.getSequenceFeatures() != null) + { + + List sfs = entry.getSequenceFeatures(); + for (SequenceFeature feature : sfs) + { + SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature) + : new SequenceFeature[] { new SequenceFeature(feature) }; + if (sf != null) + { + for (int sfi = 0; sfi < sf.length; sfi++) + { + addSequenceFeature(sf[sfi]); + } + } + } + } + + // transfer PDB entries + if (entry.getAllPDBEntries() != null) + { + Enumeration e = entry.getAllPDBEntries().elements(); + while (e.hasMoreElements()) + { + PDBEntry pdb = e.nextElement(); + addPDBId(pdb); + } + } + // transfer database references + DBRefEntry[] entryRefs = entry.getDBRefs(); + if (entryRefs != null) + { + for (int r = 0; r < entryRefs.length; r++) + { + DBRefEntry newref = new DBRefEntry(entryRefs[r]); + if (newref.getMap() != null && mp != null) + { + // remap ref using our local mapping + } + // we also assume all version string setting is done by dbSourceProxy + /* + * if (!newref.getSource().equalsIgnoreCase(dbSource)) { + * newref.setSource(dbSource); } + */ + addDBRef(newref); + } + } + } + + /** + * @return The index (zero-based) on this sequence in the MSA. It returns + * {@code -1} if this information is not available. + */ + @Override + public int getIndex() + { + return index; + } + + /** + * Defines the position of this sequence in the MSA. Use the value {@code -1} + * if this information is undefined. + * + * @param The + * position for this sequence. This value is zero-based (zero for + * this first sequence) + */ + @Override + public void setIndex(int value) + { + index = value; + } + + @Override + public void setRNA(RNA r) + { + rna = r; + } + + @Override + public RNA getRNA() + { + return rna; + } + + @Override + public List getAlignmentAnnotations(String calcId, + String label) + { + List result = new ArrayList(); + if (this.annotation != null) + { + for (AlignmentAnnotation ann : annotation) + { + if (ann.calcId != null && ann.calcId.equals(calcId) + && ann.label != null && ann.label.equals(label)) + { + result.add(ann); + } + } + } + return result; + } + + @Override + public String toString() + { + return getDisplayId(false); + } + + @Override + public PDBEntry getPDBEntry(String pdbIdStr) + { + if (getDatasetSequence() != null) + { + return getDatasetSequence().getPDBEntry(pdbIdStr); + } + if (pdbIds == null) + { + return null; + } + List entries = getAllPDBEntries(); + for (PDBEntry entry : entries) + { + if (entry.getId().equalsIgnoreCase(pdbIdStr)) + { + return entry; + } + } + return null; + } + + @Override + public List getPrimaryDBRefs() + { + if (datasetSequence != null) + { + return datasetSequence.getPrimaryDBRefs(); + } + if (dbrefs == null || dbrefs.length == 0) + { + return Collections.emptyList(); + } + synchronized (dbrefs) + { + List primaries = new ArrayList(); + DBRefEntry[] tmp = new DBRefEntry[1]; + for (DBRefEntry ref : dbrefs) + { + if (!ref.isPrimaryCandidate()) + { + continue; + } + if (ref.hasMap()) + { + MapList mp = ref.getMap().getMap(); + if (mp.getFromLowest() > start || mp.getFromHighest() < end) + { + // map only involves a subsequence, so cannot be primary + continue; + } + } + // whilst it looks like it is a primary ref, we also sanity check type + if (DBRefUtils.getCanonicalName(DBRefSource.PDB).equals( + DBRefUtils.getCanonicalName(ref.getSource()))) + { + // PDB dbrefs imply there should be a PDBEntry associated + // TODO: tighten PDB dbrefs + // formally imply Jalview has actually downloaded and + // parsed the pdb file. That means there should be a cached file + // handle on the PDBEntry, and a real mapping between sequence and + // extracted sequence from PDB file + PDBEntry pdbentry = getPDBEntry(ref.getAccessionId()); + if (pdbentry != null && pdbentry.getFile() != null) + { + primaries.add(ref); + } + continue; + } + // check standard protein or dna sources + tmp[0] = ref; + DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp); + if (res != null && res[0] == tmp[0]) + { + primaries.add(ref); + continue; + } + } + return primaries; + } + } + + /** + * {@inheritDoc} + */ + @Override + public List findFeatures(int fromColumn, int toColumn, + String... types) + { + int startPos = findPosition(fromColumn - 1); // convert base 1 to base 0 + int endPos = fromColumn == toColumn ? startPos + : findPosition(toColumn - 1); + + List result = getFeatures().findFeatures(startPos, + endPos, types); + + /* + * if end column is gapped, endPos may be to the right, + * and we may have included adjacent or enclosing features; + * remove any that are not enclosing, non-contact features + */ + if (endPos > this.end || Comparison.isGap(sequence[toColumn - 1])) + { + ListIterator it = result.listIterator(); + while (it.hasNext()) + { + SequenceFeature sf = it.next(); + int sfBegin = sf.getBegin(); + int sfEnd = sf.getEnd(); + int featureStartColumn = findIndex(sfBegin); + if (featureStartColumn > toColumn) + { + it.remove(); + } + else if (featureStartColumn < fromColumn) + { + int featureEndColumn = sfEnd == sfBegin ? featureStartColumn + : findIndex(sfEnd); + if (featureEndColumn < fromColumn) + { + it.remove(); + } + else if (featureEndColumn > toColumn && sf.isContactFeature()) + { + /* + * remove an enclosing feature if it is a contact feature + */ + it.remove(); + } + } + } + } + + return result; + } + + /** + * Invalidates any stale cursors (forcing recalculation) by incrementing the + * token that has to match the one presented by the cursor + */ + @Override + public void sequenceChanged() + { + changeCount++; + } + + /** + * {@inheritDoc} + */ + @Override + public int replace(char c1, char c2) + { + if (c1 == c2) + { + return 0; + } + int count = 0; + synchronized (sequence) + { + for (int c = 0; c < sequence.length; c++) + { + if (sequence[c] == c1) + { + sequence[c] = c2; + count++; + } + } + } + if (count > 0) + { + sequenceChanged(); + } + + return count; + } +}